This paper proposes a novel approach to constituting all the feasible ways of folding, based on crease infomation obtained from an image of illustrations of general origami drill books. Since crease information from 2D plane figures is superficial and incomplete, how to only generate all the feasible ways is a problem. This paper deals with this subject. Origami operations can be classified into basic operations and complex operations. For the basic ones, we propose a method that can create creases corresponding to the operations on a sheet of extended paper called unfolded plan, and an algorithm, called origamisection method, that is used to test physical feasibility of basic operations. For the complex ones, we propose several algorithms to produce correct creases for each operation, which keep the consistencyof the crease patterns under some geometrical constrains. Some experimental results are given to show the practical efficacy of the proposed methods.